Over the past three decades, research supported by this grant has had a major impact on our understanding of how cells respond to damage to their DNA. Many of our findings have been relevant to cancer. It has made particularly important contributions to our knowledge of the crucial role that translesion DNA polymerases play in DNA damage tolerance and mutagenesis and of SOS responses to DNA damage. The proposed research addresses critical problems concerning the roles of translesion DNA polymerases, mechanisms of DNA damage removal, and cell responses to DNA problems. We have discovered that Y Family DinB-related DNA polymerases - E. coli DinB (DNA pol IV), its mammalian ortholog Pol kappa, and its archaeal orthologs - have a striking ability to preferentially carry out accurate translesion synthesis (TLS) over certain adducts typified by N2-furfuryl-dG. We have termed these stealth lesions since the high ratio of DinB to replicative polymerase makes them largely invisible to DNA replication, but they remain in the DNA to cause problems with transcription. The proposed experiments will yield new detailed insights into the properties of DinB that endow it with this special characteristic and test whether the associated costs are DinB's propensities to -1 nt deletion mutations and to incorporate oxidized nucleotides in a mutagenic fashion. The experiments will also offer fresh insights into the complex regulatory processes that regulate access of potentially mutagenic TLS DNA polymerases to termini. We have discovered unanticipated roles in DNA repair and damage tolerance for NusA, a component of elongating RNAP polymerases. We have proposed a previously unrecognized pathway of NusA-dependent transcription-coupled repair (TCR) that is of particular importance for the removal of stealth lesions from the transcribed strand of expressed genes. We have also proposed a model for NusA- dependent transcription-coupled TLS (TC-TLS), the first for any organism, that can help cells deal with transcriptional problems created by gaps in the transcribed strand that result from lesions in the non transcribed strand. Our experiments will define the relationship between a lesion's ability to be preferentially bypassed by DinB, to block transcription, and to be recognized by nucleotide excision repair (NER). They will also provide in vivo and in vitro tests of our models of NusA-dependent TCR and TC-TLS. Hydroxyurea (HU), an inhibitor of class I ribonucleotide reductases, is widely used to block DNA replication. We made the unexpected discoveries that cells die after HU treatment because a sequence of cellular events results in the production of toxic hydroxyl radicals and that alteration of a TLS DNA polymerase can prevent the lethality associated with HU treatment. The proposed experiments will offer key insights into the molecular details of the processes that underlie these phenomena and could lead to the identification of new drug targets.